1. Field of the Invention
The present invention relates to circuitry for generation of periodic signals such as dock signals. More specifically, the present invention relates to a delay line circuit for register controlled digital delay locked loop (DDLL) circuits which use fewer gates and have improved performance.
2. Description of the Related Art
Many high speed electronic systems possess critical timing requirements which dictate the need to generate a periodic clock wave form that possesses a precise time relationship with respect to some reference signal. The improved performance of computing integrated circuits (ICs) and the growing trend to include several computing devices on the same board present a challenge with respect to synchronizing the time frames of all the components.
While the operation of all components in the system should be highly synchronized, i.e., the maximum skew or difference in time between the significant edges of the internally generated clocks of all the components should be minute, it is not enough to feed the reference clock of the system to all the components. This is because different chips may have different manufacturing parameters which, when taken together with additional factors such as ambient temperature, voltage, and processing variations, may lead to large differences in the phases of the respective chip generated clocks.
Conventionally synchronization is achieved by using DDLL circuits to detect the phase difference between clock signals of the same frequency and produce a digital signal related to the phase difference. By feeding back the phase difference-related signal to control a delay line, the timing of one clock signal is advanced or delayed until its rising edge is coincident with the rising edge of a second clock signal.
The operation of conventional DDLLs is shown in FIGS. 1 and 2. In FIG. 1, clock input buffer 104, delay lines 101, 102, and data output buffer 109 constitute an internal clock path. Delay line 101 is a variable delay generator with a logic-gate chain. A second delay line 102 is connected to replica circuits 108, which emulate the internal clock path components. Replica circuits 108 include dummy output buffer 110, with dummy load capacitance 111 and dummy clock buffer 107. The dummy components and second delay line 102 constitute a dummy clock path having exactly the same delay time as the internal clock path. Shift register 103 is used for activating a number of delay elements in both delay lines based on a command generated by phase comparator 106.
Phase comparator 106 compares the dummyclock and the external clock phases which differ by one cycle. This comparison is illustrated in FIGS. 2A, 2B, 2C, and 2D. External dock signal 200 is divided down in divider 105 to produce divided-down external signal 201. Signal 202 is the signal at the output of dummy delay line 102. Signal 203, which is generated inside phase comparator 106, is a one delay unit delayed output dummy line signal 202. If both signals 202 and 203 go high before 201 goes low, this means that the output clock is too fast and phase comparator 106 outputs a shift left (SL) command to shift register 103, as illustrated in FIG. 2B. Shift register 103 shift the tap point of delay lines 102 and 101 by one step to the left, increasing the delay. Conversely, if both signals 202 and 203 go high after 201 goes low, this means that the output dock is too slow and phase comparator 106 outputs a shift right (SR) command to shift register 103, as illustrated in FIG. 2D. Shift register 103 shifts the tap point of delay lines 102 and 101 by one step to the right, decreasing the delay. If 201 goes low between the time 202 and 203 go high, the internal cycle time is properly adjusted and no shift command is generated, as illustrated in FIG. 2C. The output of the internal clock path in this case coincides with the rising edge of the external dock and is independent of external factors such as ambient temperature and processing parameters.
A schematic diagram of a conventional Vernier Delay Line (VDL) circuit 300 used for the stages of delay line 101 of FIG. 1 is shown in FIG. 3. The circuit 300 of FIG. 3 consists of a series of n delay stages, each stage consisting of three NAND gates 305, 306 and 307 and two inverters 310, 311. The unit delay for stage 301 of upper delay line 302 consists of NAND gate 305 and inverter 310. The upper delay line 302 and lower delay line 303 are connected through NAND switch 306 whose transistor gates become the load for the upper delay line 302. Shift register 315 provides a signal to open or close NAND switch 306. The delay of the upper delay line 302 slightly exceeds that of the lower delay line 303. This delay difference becomes the unit delay time of the VDL circuit 300.
FIG. 3A illustrates in block diagram form the functioning of the VDL circuit 300 of FIG. 3 Each unit delay 350, 351, 352, 353, 354 in upper delay line 302 has a delay time of 1.2 td, and each unit delay 360, 361, 362, 363, 36,4 in lower delay line 303 has a delay time of td, where td is the unit delay time of the conventional delay generator. The additional 0.2 td delay of the upper delay line in this example is due to the gate loading from the NAND switches. These unit delays 350-354 and 360-364 are serially connected through switches 370, 371, 372, 373, and 374. If only switch 370 doses, the VDL generates a delay of 5 td from IN node 340 to OUT node 399. Similarly, if switch 371 closes, the VDL generates a delay time of 5.2 td from IN node 340 to OUT node 399. Thus, the VDL circuit 300 is capable of generating a delay of every 0.2 td delay time.
Conventional delay lines of DDLLs, however, suffer from numerous drawbacks. One such drawback is that the resolution, i.e., the delay per stage, of the delay line is dependent upon the number of gates for each unit delay of the stage. The larger the number of gates in each unit delay, the larger the unit delay time td. Although the circuit shown in FIG. 3 can generate a delay of every 0.2 td, the resolution is limited by the value of td. The larger the value of td, the lower the resolution possible.
In addition to providing poor resolution, a high value for the unit delay time td can cause problems when the DDLL is placed in a state of minimum delay. A state of minimum delay occurs when the delay between the input and output clock signals is as close to zero as allowed by the parameters of the delay line, i.e., the smallest delay as allowed by the unit delay time td. In this case, if the DDLL attempts to decrease the delay, such decrease would be impossible because the delay line is already at minimum delay. Each unit delay of the delay line shown in FIG. 3 consists of one NAND gate and one inverter. The unit delay time for each unit delay having this construction is approximately 200-300 picoseconds. The minimal delay of the delay line 300 is thus limited to 200-300 picoseconds, without the possibility of decreasing the unit delay time below that time. Thus, the resolution of the delay line, determined by the unit delay time, is limited by the number of gates in each unit delay. is A further drawback of conventional DDLL circuits is tile space required to layout the circuitry of the DDLLs. Each stage of the delay line consists of three NAND gates and two inverters for a total of five gates. Each stage could be replicated 50-100 times to target a typical clock input frequency of 100 MHz. This extensive amount of circuitry occupies a significant amount of space within a semiconductor circuit.
Yet another drawback of conventional DDLL circuits is that they are inherently inaccurate due to asymmetries in the delay line design. Each stage of the delay line consists of three NAND gates and two inverters . Unless the pull-up and pull-down times of the transistors forming the inverters and NANDs in each delay element are identical, the output of the delay line will consist of pulses with asymmetrical rising and falling edges as compared to the input signal. This asymmetry leads to differing pulse widths between the input signal and the output signal, as shown in FIG. 5A. The output signal, therefore, will differ in pulse width from the input signal, which may lead to inaccuracies.
There is a need, therefore, to improve the performance of the delay line in a DDLL circuits by increasing the resolution of the delay line. Additionally, there is a need for improving the configuration of the delay line in DDLL circuits to reduce the amount of space required for the circuitry used to implement the delay line.
The present invention provides a unique method and apparatus for improving the resolution of a delay line, while also substantially reducing the necessary circuitry and associated space required for layout by reducing the number of gates in each unit delay.
In accordance with the present invention, the gate count for each unit delay is reduced to one gate. Since the number of gates for each unit delay is minimal, the unit delay time is decreased to a minimum, improving the resolution of the delay line.
Furthermore, by reducing the number of gates for each stage of the delay line to a total of three gates (two NAND gates and one inverter), the delay line will occupy approximately 40% less of the area previously occupied by the conventional delay line.
Finally, by reducing the number of gates for each stage of the delay line to a total of three gates, the delay line will result in substantially symmetrical rising and failing edges of the output signal.
These and other advantages and features of the invention will become apparent from the following detailed description of the invention which is provided in connection with the accompanying drawings.